Effect of Polyethylene and Steel Fibers on the Fracture Behavior of Coral Sand Ultra-High Performance Concrete

Abstract

As a representative high-performance construction material, ultra-high performance concrete (UHPC) is typically prepared using quartz sand and steel fibers. To alleviate the shortage of building materials in island and reef regions, this study employs coral sand for UHPC preparation and investigates the effects of different fibers on its mechanical properties. This study demonstrates that this approach mitigates brittle failure patterns and enhances the durability of structures. To investigate the enhancement effects of PE and steel fibers on the mechanical properties of coral sand ultra-high performance concrete (CSUHPC), 12 mix designs were formulated, including a plain (no fiber) reference group and PE fiber-reinforced, steel fiber-reinforced, and hybrid fiber combinations. Compressive tests, tensile tests, and three-point bending tests on pre-notched beams were conducted. Key parameters such as 28-day compressive strength, tensile strength, and flexural strength and toughness were measured. A multi-criteria evaluation framework was established to comprehensively assess the integrated performance of each group. The experimental results demonstrated that fiber incorporation significantly enhanced the compressive strength and fracture properties of CSUHPC compared to the plain reference group. Steel fiber-only reinforcement exhibited the most pronounced improvement in compressive strength and fracture properties, while hybrid fiber combinations provided superior tensile performance. Through the established multi-criteria evaluation framework, the optimal comprehensive performance was achieved with a 3% steel fiber dosage, achieving improvements of 0.93 times in compressive strength, 2.80 times in tensile strength, 1.84 times in flexural strength, 192.08 times in fracture energy, and 1.84 times in fracture toughness relative to the control group.

1. Introduction

In recent years, offshore island reef construction has progressively advanced to a stage characterized by structural reinforcement of existing buildings and high-performance structural design of new infrastructure. Nevertheless, construction projects on island reefs consistently face challenges due to resource scarcity [1,2]. When conventional aggregates are utilized, marine transportation is required, resulting in elevated costs, extended logistics cycles, and material wastage [3,4]. To address this, researchers have explored the use of marine-sourced materials such as coral aggregates as substitutes for conventional aggregates in concrete production [5,6,7]. On some islands in the Pacific Ocean, infrastructure such as airports and roads constructed by the U.S. using coral concrete has remained in service since World War II [8]. However, the inherent characteristics of coral aggregates—particularly their high porosity and low uniaxial compressive strength—adversely affect concrete performance [9,10,11,12]. To produce high-strength coral aggregate concrete (CAC) with enhanced properties, researchers have optimized the concrete’s internal pore structure through gradation adjustment of coral aggregates and fiber dosage optimization [13,14,15,16].

The strength enhancement of concrete exacerbates its brittle behavior, resulting in more severe cracking. Coral aggregate concrete exhibits significantly higher brittleness than conventional concrete [17]. necessitating rigorous investigation of its fracture characteristics. Han et al. studied the fracture properties of CAC with varying strength grades, revealing that increased strength reduces the area under the load-crack mouth opening displacement (P-CMOD) curve, indicating a diminished energy absorption capacity [18]. Wang investigated the effects of steel fiber (SF) and polyvinyl alcohol fiber (PVA) reinforcement on the mechanical properties of CAC, demonstrating that fiber addition mitigates strength reductions caused by inherent defects in coral aggregates [19]. Dong reported that polypropylene fiber (PPF) incorporation enhances the flexural toughness of CAC [20]. Wang further examined the influence of fiber dosage variations on the dynamic mechanical properties of CAC, confirming a significant toughness improvement through fiber reinforcement [21]. Current research primarily focuses on conventional CAC, while studies on coral aggregate ultra-high performance concrete (CSUHPC) remain limited.

UHPC exhibits low permeability and high compactness due to its dense matrix, effectively mitigating external environmental erosion while demonstrating superior strength and durability [22,23,24]. Existing research demonstrates that the fiber bridging effect remains highly effective even in large-span reinforced concrete beams [25]. In tropical marine environments, such as island reefs with high-humidity/high-salt exposure, conventional concrete suffers accelerated durability degradation from corrosive ion ingress [26,27]. Under such harsh service conditions, UHPC maintains exceptional performance, showing significant potential for marine infrastructure applications [28,29]. While steel fiber reinforcement significantly enhances concrete’s strength–toughness properties, its susceptibility to chloride-induced corrosion in marine environments limits offshore applications. Conversely, polyethylene (PE) fibers demonstrate superior chemical inertness and corrosion resistance, substantially improving concrete durability in aggressive conditions [30,31]. To synergistically enhance the mechanical–durability performance of CSUHPC, this study employs corrosion-resistant copper-coated steel fibers and ultra-high-molecular-weight polyethylene fibers, systematically investigating the fiber dosage effect on CSUHPC’s strength–toughness enhancement.

In this paper, compressive strength, tensile strength, and pre-notched beam tests were conducted on CSUHPC with different fiber volume fractions to explore the effects of fiber content on the mechanical properties and toughness of CSUHPC. Additionally, a multi-criteria evaluation system was established to determine the optimal fiber volume fraction for CSUHPC.

2. Materials and Methods

2.1. Raw Materials

The primary binder materials employed in formulating CSUHPC were Type II Portland cement (Grade 52.5), silica fume, and glass microspheres, all sourced from Nanjing, China. Coral sand, sourced from a tropical island and processed through mechanical sieving, exhibited a maximum particle size of 2.5 mm with a fineness modulus of 2.4. The primary chemical constituents of the coral sand are presented in

Table 1. Chemical composition table of coral sand/%.

Ingredient	CaO	SiO2	Al2O3	Fe2O3	SrO	MgO	Na2O	SO3	K2O	TiO2
Content	85.25	6.06	2.43	2.07	1.91	0.50	0.46	0.43	0.35	0.29

, while its morphological characteristics are illustrated in Figure 1. The coral sand gradation is composed as follows: 20% of particles ranging from 0 to 0.315 mm, 35% from 0.315 to 0.63 mm, 35% from 0.63 to 1.25 mm, and 10% from 1.25 to 2.5 mm. PE fibers manufactured in Beijing and steel fibers produced in Henan Province were incorporated. Their mechanical properties are detailed in

Table 2. Different fiber physical properties.

Fiber Type	Length/mm	Diameter/mm	Density/g·cm−3	Tensile Strength/MPa
Steel Fiber	12	0.22	7.85	3000
PE Fiber	12	0.0156	0.97	3980

. A polycarboxylate-based high-range water reducer and silicone defoamer, both sourced from Tianjin, were utilized at dosages of 1.5% and 0.15% by mass of cementitious materials, respectively.

2.2. Mix Proportions

Based on the primary fiber volume content, hybridizing with a second fiber type could combine the characteristics of both fibers and create synergistic effects on the mechanical properties of concrete [32]. Preliminary tests indicated that both the workability and mechanical performance of CSUHPC significantly deteriorated when the total fiber volume content exceeded 3%. Therefore, this study utilized hybrid steel–PE fibers while maintaining a total fiber volume content below 3%. To investigate the influence of different fiber dosages on CSUHPC properties, experimental groups were designed, including a control group (0% fiber), three single PE fiber groups (0.5%, 1%, and 1.5% volume fractions), three single steel fiber groups (1%, 2%, and 3% volume fractions), and five hybrid fiber combinations; the mix proportions of CSUHPC are detailed in

Table 3. Mix proportions (mass ratio).

Series	Coral Sand	Cement	Glass Microspheres	Silica Fume	Water	Defoamer	Water Reducer	PE Fiber	Steel Fiber
P0G0	1	0.7	0.15	0.15	0.23	0.0015	0.015	0	0
P0.5G0	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.005	0
P1G0	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.010	0
P1.5G0	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.015	0
P0G1	1	0.7	0.15	0.15	0.23	0.0015	0.015	0	0.078
P0G2	1	0.7	0.15	0.15	0.23	0.0015	0.015	0	0.156
P0G3	1	0.7	0.15	0.15	0.23	0.0015	0.015	0	0.234
P0.5G1	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.005	0.078
P1G1	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.010	0.078
P1.5G1	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.015	0.078
P0.5G2	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.005	0.156
P1G2	1	0.7	0.15	0.15	0.23	0.0015	0.015	0.010	0.156

.

2.3. Preparation Process of CSUHPC

Step 1: Coral sand, silica fume, glass microspheres, cement, and defoamer were dry-mixed in a forced-action mixer for 3 min until the powder components achieved visual homogeneity. Step 2: The water reducer was dissolved in the mixing water prior to batching. This solution was then added incrementally to the mixer during low-speed mixing in two separate batches, with each addition followed by 3 min of continuous mixing. Step 3: After the slurry was homogenized, PE fibers and steel fibers were incorporated while stirring. Step 4: Mixing continued for an additional 3 min following fiber addition. Step 5: The CSUHPC slurry was molded, vibrated, and cured under standard conditions for 28 days. The specimens were cured at 20 ± 2 °C and ≥95% relative humidity.

Three replicate specimens were cast for each mix proportion, and the average value of the test results was adopted.

2.4. Compressive Strength and Splitting Tensile Strength Tests of CSUHPC

The CSUHPC compressive strength test was conducted using 100 mm cube specimens with a loading rate of 10 kN/s [33], and the loading device is shown in Figure 2. The CSUHPC tensile strength test was conducted using dog bone specimens with a loading rate of 0.1 mm/min [34], and the loading device diagram is shown in Figure 3.

2.5. CSUHPC Bending Performance Test

To test the fracture properties of CSUHPC, three-point bending tests were performed using pre-cut specimens. The dimensions of the beam were L = 400 mm, t = 100 mm, and H = 100 mm (rectangular), respectively, and a deep cut of a₀ = 10 mm reserved in the middle of the bottom of the specimen. The spacing between the two pivot points was S = 350 mm. The dimensions of the specimen are shown in Figure 4a below. A displacement-controlled loading rate of 0.1 mm/min was applied throughout testing. The CMOD was measured using a clip-type extensometer, while the universal testing machine simultaneously recorded load–displacement data. The bending test configuration is shown in Figure 4b.

Flexural strength denotes concrete’s fracture resistance in three-point bending, while residual strength represents its post-failure load-bearing capacity. Per residual strength criteria in [35], this study defines the strength at 5 mm crack mouth opening displacement (CMOD) on the P-CMOD curve as specimen residual strength. Plain CSUHPC specimens (P0G0) exhibited brittle fracture at initial cracking (CMOD = 0.19 mm), hence their residual strength was excluded. The flexural strength calculation formula is as follows:

$$f_f={\displaystyle \frac{3PS}{2t{(H-a_0)}^2}}$$

(1)

In the formula: f_f is the flexural strength of the CSUHPC specimen. P is the load. S is the span between the two supports of the base. t is the thickness of the specimen. H is the height of the specimen. a₀ is the initial crack height.

Fracture energy (G_F) represents the energy absorbed per unit area during fracture propagation in CSUHPC specimens [36]. The ductility index (D_u) quantifies deformation resistance capacity [37].

$$G_F={\displaystyle \frac{W_0+mg{\delta }_{\mathit{max}}}{t(H-a_0)}}$$

(2)

$$D_u={\displaystyle \frac{G_F}{P_{\mathit{max}}}}$$

(3)

In the equation, W₀ represents the area enclosed by the P-CMOD curve and the coordinate axes. m is the mass between the supports of the specimen. g is the acceleration due to gravity. δ_max is the displacement at the loading point when the specimen fails. P_max is the peak load of the specimen.

Fracture toughness (K_IC) is the stress intensity factor corresponding to the crack propagation in concrete. It is calculated according to the method recommended by ASTM in the United States, as follows [38]:

$$K_{IC}={\displaystyle \frac{P_{\mathit{max}}S}{t{H}^{\frac{3}{2}}}}f(a)$$

(4)

$$f(a)=2.9a^{\frac{1}{2}}-4.6a^{\frac{3}{2}}+21.8a^{\frac{5}{2}}-37.6a^{\frac{7}{2}}+38.7a^{\frac{9}{2}}$$

(5)

$$a={\displaystyle \frac{a_0}{H}}$$

(6)

In the equation, f_(a) is a function related to the initial crack height. a is the crack height ratio.

2.6. Strength Projections and Comprehensive Performance Evaluation

As per composite mechanics theory [39,40], CSUHPC strength comprises matrix strength and fiber contribution strength. Accounting for fiber–matrix bond strength and hybrid fiber synergy effects, the fiber strength component incorporates an enhancement coefficient (k) and an empirical constant (C). The theoretical model is expressed as follows:

$$f=f_b+k_1{x}_1+k_2{x}_2+C$$

(7)

In the equation, f represents the strength of CSUHPC. f_b is the matrix strength. x₁ and x₂ are the volume fractions of PE fibers and steel fibers, respectively. k₁ and k₂ are the enhancement coefficients. And C is the constant term.

The area S_i and perimeter C_i of the radar chart can be used as features of the radar chart. At the same time, evaluation vectors (A_i, B_i) are constructed, where A_i represents the degree of excellence of the specimen in terms of mechanical properties, and B_i represents the degree of balance in the specimen’s mechanical properties. The F_i represents the overall performance excellence of the specimen. The calculation methods for these values are referenced in the literature [41].

$$S_i={\displaystyle \sum _{j=1}^m\frac{n_{ij}{n}_{i(j+1)}\sin \theta }{2}}$$

(8)

$$C_i={\displaystyle \sum _{j=1}^m\sqrt{n_{ij}^2+n_{i(j+1)}^2-2n_{ij}{n}_{i(j+1)}\cos \theta }}$$

(9)

$$A_i={\displaystyle \frac{S_i}{\max \left\{S_i\right\}}}$$

(10)

$$B_i={\displaystyle \frac{4\pi S_i}{C_i^2}}$$

(11)

$$F_i=\sqrt{A_i{B}_i}$$

(12)

3. Results and Discussion

3.1. Compressive Strength and Tensile Strength Tests

The compressive and tensile strengths of CSUHPC with varying fiber volume fractions are shown in Figure 5. Within single-fiber systems, strength increased progressively with fiber content. Compared to the plain matrix (P0G0), 1.5% PE fibers increased compressive and tensile strength by 0.63-fold and 2.70-fold, respectively, while 3% steel fibers enhanced them by 0.93-fold and 2.80-fold. Steel fibers demonstrated superior compressive strength enhancement but less significant tensile strength improvement versus PE fibers. Dispersed fibers formed dense three-dimensional networks within the matrix, effectively redistributing stresses to less loaded regions and enhancing overall strength. However, due to higher density and smoother surface morphology, steel fibers achieved superior dispersion within the CSUHPC matrix. Conversely, PE fibers’ smaller diameter and lower distribution efficiency promoted localized clustering in the cementitious system, intensifying pre-existing flaws and consequently constraining compressive strength enhancement. PE fibers exhibited significantly higher spatial density than steel fibers, forming finer three-dimensional networks. Furthermore, steel fibers’ smooth surfaces yielded inferior fiber–matrix interfacial bond strength compared to PE fibers. The selected PE fibers possessed higher elastic modulus and tensile strength, intrinsically enhancing post-cracking strain-hardening behavior. Consequently, PE fibers demonstrated greater efficacy in improving CSUHPC tensile performance relative to steel fibers.

The incorporation of PE fibers into steel fiber-reinforced mixtures further enhanced the mechanical properties of CSUHPC. When 1% steel fibers were combined with 0.5%, 1%, and 1.5% PE fibers, the compressive strength increased by 3.7%, 10.6%, and 12.4%, respectively, compared to the P0G1. Meanwhile, the tensile strength increased by 22.7%, 68.2%, and 93.2%, respectively. When 2% steel fibers were combined with PE fibers, the P0.5G2 and P1G2 of compressive strength increased by 3.0% and 4.8%, respectively, compared to the P0G2. Meanwhile, the tensile strength increased by 5.8% and 18.8%, respectively. Compared to the steel fiber-reinforced group with the same steel fiber content, the hybrid fiber group showed a less pronounced improvement in compressive strength but a more significant enhancement in tensile strength. When compared to the P0G2, the hybrid fiber group showed increases in compressive strength and tensile strength of 3.5% and 7.2%, respectively. In contrast, compared to the P0G3, the hybrid fiber group showed a slight reduction in compressive strength, while tensile strength increased by 7.9%. As shown in Figure 5, the CSUHPC developed in this study meets the minimum compressive strength requirements specified in GB/T 31387-2025 [42]. When the steel fiber content was 3%, the compressive strength of the CSUHPC reached 131.9 MPa. Steel fibers significantly enhanced CSUHPC compressive strength, while PE fibers substantially improved tensile strength. Hybrid steel–PE fiber systems yielded moderate compressive strength gains and exceptional tensile strength enhancement. At 1% steel fiber and 1.5% PE fiber volume fractions, CSUHPC tensile strength reached 8.5 Mpa—4.25 times that of the P0G0 control group. Compressive strength improvement can be ranked as follows (descending order): steel mono-fiber > hybrid > PE mono-fiber. Tensile strength improvement ranks: hybrid > PE mono-fiber > steel mono-fiber.

3.2. P-CMOD Curve

Figure 6 presents the P-CMOD curve derived from the bending test of pre-notched beams. As shown in the figure, upon fiber addition, the CSUHPC specimens transitioned from brittle to ductile failure. Additionally, with increasing fiber content, the ascending branch of the P-CMOD curve became steeper. The bending failure process of CSUHPC specimens comprises three stages: (1) Elastic stage: Load increases linearly with CMOD until matrix tensile capacity is reached, initiating microcracks at the notch tip. (2) Crack initiation stage: Following microcrack initiation, load increases non-linearly with reduced stiffness to peak load, during which microcracks coalesce and propagate into visible cracks. (3) Crack propagation stage: Macroscopic cracks propagate vertically from the notch tip, reducing the specimen’s load-bearing capacity. Plain CSUHPC (P0G0) undergoes brittle fracture when cracks reach critical length; load drops catastrophically from peak values. In fiber-reinforced specimens, fiber bridging mechanisms decelerate post-peak load decay, demonstrating enhanced toughness. As shown in Figure 7, the specimen exhibits the characteristic of “cracking without complete failure” at the moment of rupture.

As shown in Figure 6a, the PE fiber-reinforced CSUHPC curve exhibits double peak values and stress softening characteristics. With increasing fiber content, the peak load increases. After crack initiation in the specimen, the load drops rapidly from the first peak point. At this stage, the tensile stress of PE fibers at the pre-notched tip restrains crack propagation until the load reaches the second peak value. Thereafter, the bridging effect of PE fibers begins to diminish, and the specimen’s bearing capacity starts to decline. Both the steel fiber and hybrid fiber groups exhibited stress hardening characteristics, with steel fibers demonstrating a significantly stronger effect on enhancing the fracture toughness of the specimens compared to PE fibers. The failure mode of steel fibers in CSUHPC was primarily pull-out, resulting in a relatively gradual decline in the descending branch of the curve. In contrast, PE fibers exhibited better bonding with the ITZ and failed mainly by fracture or pull-out in CSUHPC specimens. When PE fibers fail, the curve shows an abrupt drop.

3.3. Flexural Strength and Residual Strength

Figure 8 shows the flexural strength and residual strength of CSUHPC. As indicated in the figure, the addition of fibers enhanced the flexural strength and residual strength of CSUHPC, and the strength increased with increasing fiber content. For single-fiber systems, compared to the plain CSUHPC, the addition of 0.5%, 1%, and 1.5% PE fibers increased the flexural strength of CSUHPC specimens by 4.5%, 10.1%, and 33.7%, respectively. The addition of 1%, 2%, and 3% steel fibers increased the flexural strength of CSUHPC specimens by 0.89, 1.48, and 1.84 times, respectively. When the CMOD reached 5 mm, the residual strength of steel fiber-reinforced specimens was significantly higher than that of PE fiber-reinforced specimens. At a fiber content of 1%, the flexural strength and residual strength of steel fiber specimens increased by 71.4% and 2.78 times, respectively, compared to those of PE fiber specimens. From the P-CMOD curve, it can be observed that the flexural strength growth rate of PE fiber specimens is high, but when the CMOD reaches a critical value, the bridging effect of PE fibers diminishes. In contrast, due to their higher stiffness, steel fibers exhibit a broader bridging zone at the pre-notched tip compared to PE fibers, leading to a more significant enhancement in CSUHPC strength.

For the hybrid fiber system, when 0.5%, 1%, and 1.5% PE fibers are added to the 1% steel fiber base, the flexural strength of hybrid fiber specimens increases by 2.4%, 13.1%, and 22.6%, respectively, compared to the P0G1. The residual strength of hybrid fiber specimens is lower than that of the 1% steel fiber group but higher than those of specimens with PE fibers alone. When 0.5% and 1% PE fibers are combined with a 2% steel fiber base, the flexural strength of hybrid fiber specimens increases by 3.6% and 6.8%, respectively, compared to the P0G2. The residual strength increases by 20.5% and 9.9%, respectively. At the same total fiber content, the flexural strength and residual strength of the steel fiber group are both higher than those of the hybrid fiber group. In the descending branch of the P-CMOD curve, agglomeration of some PE fibers within CSUHPC introduces additional inherent defects in the specimens. Additionally, when the CMOD is large, the bridging effect of PE fibers diminishes, resulting in a steeper decline in the curve than in the steel fiber group. Based on the above analysis, it can be concluded that the single steel fiber group achieves the best enhancement in flexural strength for CSUHPC, followed by the hybrid fiber group, while the single PE fiber group shows relatively limited improvement in flexural strength.

3.4. Fracture Energy and Ductility Index

Figure 9 shows the fracture energy and ductility index of CSUHPC. As indicated in Figure 9, the addition of fibers significantly improves the toughness of CSUHPC. The single steel fiber group exhibits stronger enhancement in fracture properties than both the hybrid fiber group and the single PE fiber group. The fracture energy of the 1.5% PE fiber group is 53.88 times that of the P0G0 group, while the 3% steel fiber group shows an increase to 192.08 times that of the P0G0 group. The fracture energy of the 2% steel fiber group is slightly higher than that of the 1% steel fiber group, possibly due to uneven distribution of steel fibers within the specimen, which causes a faster decline in the P-CMOD curve and a relatively smaller calculated area under the curve. In the 2% steel fiber group, the addition of 0.5% and 1% PE fibers increases the fracture energy by 12.7% and 11.5%, respectively. At the same total fiber content, the fracture energy of the 2% and 3% steel fiber groups increases by 31.8% and 23.9%, respectively, compared to the hybrid fiber group. As shown in Figure 9, the ductility index of the single PE fiber group is approximately 0.3, that of the single steel fiber group is approximately 0.5, and that of the hybrid fiber group is approximately 0.4. The fracture energy of the 2% steel fiber group is slightly higher than that of the 1% steel fiber group, its peak load is significantly higher, resulting in a lower ductility index for the 2% steel fiber group than that of the 1% steel fiber group. Based on the analysis of fracture energy and ductility index, it can be concluded that the single steel fiber group provides the best improvement in the fracture performance of CSUHPC, followed by the hybrid fiber group. When compared to the fracture energy of some UHPC specimens reported in the literature [43,44], the CSUHPC developed in this study exhibits higher fracture energy.

The P0G0 curve exhibits a lower peak load and lacks a descending branch, resulting in its fracture energy being significantly smaller than that of the fiber-reinforced specimens. In summary, the peak load of PE fiber specimens is lower than that of the steel fiber group, but their load increase rate is faster, making them suitable for the micro-crack initiation stage. Steel fibers significantly enhance the strength of CSUHPC and exhibit a wide bridging range, effectively resisting crack propagation throughout the entire testing process. The hybrid incorporation of PE fibers and steel fibers can mitigate the abrupt drop in load-bearing capacity caused by PE fiber fracture. However, the addition of PE fibers increases inherent defects in CSUHPC, leading to less significant strength improvement compared to the single steel fiber group.

3.5. Fracture Toughness

P0G0 is considered the baseline group for fracture toughness, and the growth multiples of fracture toughness for different fiber contents are plotted. As indicated in Figure 10, the addition of fibers enhances the fracture toughness of CSUHPC. For single-fiber systems, the fracture toughness increases with increasing fiber content of PE fibers and steel fibers. Compared to the control group, the fracture toughness of CSUHPC specimens increases by 34.1% with the addition of 1.5% PE fibers. The P0G3 of fracture toughness of CSUHPC specimens increases to 1.84 times that of the control group. For the hybrid fiber system, incorporating PE fibers into steel fiber-reinforced CSUHPC can further enhance the fracture toughness of the specimens. Using 1% steel fibers as the base, adding 0.5%, 1%, and 1.5% PE fibers results in increases in fracture toughness of 2.5%, 13.3%, and 22.7%, respectively. Using 2% steel fibers as the base, adding 0.5% and 1% PE fibers results in increases in fracture toughness of 3.6% and 6.7%, respectively. It can be observed that steel fiber alone provides the best improvement in fracture toughness, followed by the hybrid fiber system and then PE fiber alone.

3.6. CSUHPC Strength Projections and Comprehensive Performance Evaluation

3.6.1. CSUHPC Strength Projections

By substituting the experimental values into Equation (7),

Table 4. Strength regression equations for CSUHPC.

	Fiber Type	Regression Equations	R2
Compressive strength	PE	f = 11.02 x1 + 93.62	0.9297
	Steel	f = 8.98 x2 + 104.53	0.9943
	Hybrid	f = 8.80 x1 + 4.66 x2 + 110.57	0.8607
Tensile strength	PE	f = 2.10 x1 + 4.27	0.9992
	Steel	f = 1.60 x2 + 3.10	0.9046
	Hybrid	f = 2.84 x1 + 1.36 x2 + 2.90	0.9481
Flexural strength	PE	f = 2.60 x1 + 7.73	0.8879
	Steel	f = 4.25 x2 + 12.90	0.9801
	Hybrid	f = 3.00 x1 + 5.07 x2 + 10.87	0.9857

is generated. The R² values in Table 4 are all greater than 0.85, indicating that the equation provides a good prediction of the overall strength. Notably, the R² for tensile strength exceeds 0.9, demonstrating high prediction accuracy. The predicted compressive and flexural strengths for the PE fiber group are relatively low, possibly due to PE fiber aggregation within the specimens, which increases inherent defects and results in a relatively low fitting accuracy.

3.6.2. Comprehensive Performance Evaluation

To systematically evaluate the comprehensive performance of CSUHPC specimens, a performance evaluation system based on the radar chart method was established, with compressive strength, tensile strength, flexural strength, fracture energy, and fracture toughness as the core parameters. Due to significant numerical differences among parameters, the data were first normalized. After data processing, a radar chart was plotted, as shown in Figure 11.

The evaluation results of CSUHPC with different fiber contents was obtained using the above calculation method, as shown in

Table 5. Multi-criteria performance assessment of CSUHPC using radar chart visualization.

Series	Si	Ci	Ai	Bi	Fi
P0G0	0.2046	2.0229	0.0899	0.6281	0.2376
P0.5G0	0.5103	2.9375	0.2241	0.7428	0.4080
P1G0	0.6430	3.2286	0.2824	0.7747	0.4677
P1.5G0	0.8418	3.6959	0.3697	0.7740	0.5350
P0G1	1.0913	4.0929	0.4793	0.8182	0.6262
P0G2	1.6917	4.9714	0.7430	0.8597	0.7992
P0G3	2.2768	5.7600	1.0000	0.8619	0.9284
P0.5G1	1.1252	4.0983	0.4942	0.8414	0.6449
P1G1	1.4424	4.6438	0.6335	0.8401	0.7295
P1.5G1	1.6836	5.0263	0.7394	0.8370	0.7867
P0.5G2	1.8794	5.2317	0.8255	0.8624	0.8437
P1G2	2.0224	5.4351	0.8883	0.8599	0.8740

. By comparing the comprehensive performance indices (F_i values), we observed that the P0G3 group exhibited the best overall performance, followed by P1G2. As indicated in Table 5, in the single-fiber system, the overall performance of CSUHPC specimens improved with increasing fiber content, but PE fibers exhibited a weaker improvement effect than steel fibers. Steel fibers, characterized by high density and high elastic modulus, distributed more uniformly during the specimen preparation process, thereby significantly enhancing the overall performance of CSUHPC. Through hybrid fiber reinforcement, the overall performance of the specimens was further enhanced. For P0G1, the overall performance of hybrid fiber specimens was superior to that of single 1% steel fiber specimens. For P0G2, the overall performance of hybrid fiber specimens was superior to that of single 2% steel fiber specimens but inferior to that of P0G3. The primary issue was that the low density of PE fibers led to poor dispersion during the mixing process, which consequently amplified the inherent flaws in CSUHPC.

4. Conclusions

To address the significant degradation in strength and durability of concrete in marine environments, this study utilized coral aggregate for the production of UHPC. By incorporating both steel fibers and PE fibers, the strength, toughness, and durability of CSUHPC were further enhanced. In this paper, compressive, tensile, and pre-notched beam tests were conducted on CSUHPC specimens with different fiber volume fractions to investigate the effects of varying fiber volume fractions on the mechanical properties and toughness of CSUHPC. Additionally, a multi-criteria evaluation system was established to determine the optimal fiber volume content for CSUHPC. The CSUHPC developed in this study demonstrated superior overall performance compared to conventional concrete, making it suitable for construction projects on tropical islands and reefs.

(1)

The incorporation of fibers into CSUHPC specimens significantly improved their compressive strength. Adding steel fibers alone yielded the most significant increase in compressive strength, followed by the hybrid fiber group, while PE fibers resulted in a relatively limited enhancement in compressive strength. When the steel fiber content was 3%, the compressive strength of the CSUHPC specimen reached 131.9 MPa, representing a 93% increase over the control group.

(2)

When the PE fiber content was 1.5% and the steel fiber content was 1%, the tensile strength of CSUHPC reached a peak value of 8.5 MPa, 4.25 times higher than the control group. A comparison of the tensile strength across different specimen groups revealed that the hybrid fiber system provided the best improvement, while single steel fibers exhibited a relatively less significant enhancement.

(3)

The P-CMOD curve showed that PE fiber-reinforced CSUHPC specimens exhibited stress-softening behavior, while the steel fiber and hybrid fiber groups exhibited stress-hardening behavior. After fiber addition, the failure mode of the specimens shifted from brittle to ductile. Based on a comparison of fracture parameters, the effectiveness of fiber reinforcement for CSUHPC specimens was ranked as follows: steel fibers > hybrid fibers > PE fibers. The specimen with 3% steel fiber content demonstrated the best fracture performance, with its flexural strength, fracture energy, and fracture toughness increasing to 1.84 times, 192.08 times, and 1.84 times that of the control group, respectively.

(4)

A multi-criteria evaluation system was established to assess the overall performance of CSUHPC specimens with varying fiber contents. The optimal fiber content combination was determined to be P0G3, followed by P1G2.

This study had several limitations that merit further investigation in future research within the region. Recent rapid advances in computer technology, including machine learning and simulation methods, can be employed to model the failure process of concrete and gain deeper insights into how fiber distribution affects the mechanical behavior of CSUHPC.